synergymus Posted July 25, 2005 Posted July 25, 2005 Has anyone seen this error on thier Linux Box? [11:43:21] Completed 167500 out of 250000 steps (67%) [12:21:24] CoreStatus = 0 (0) [12:21:24] Client-core communications error: ERROR 0x0 [12:21:24] Deleting current work unit & continuing... I lost a WU with this. I hope it doesn't continue! Share this post Link to post Share on other sites More sharing options...
ExtraPickles Posted July 25, 2005 Posted July 25, 2005 Is it overclocked? I had a similar problem with a WinXP box when I popped a new chip in it last week and overclocked the balls off it and started folding with the -forceasm switch without doing much more stabilization testing beyond getting it to boot and checking load temps. Lowered the speed some and it's been chugging along great ever since. One of these days I'll get around to setting it up and testing it properly but it's main function is folding and it's doing it. Anyhow if your box is overclocked and it's stable I wouldn't think it would get errors but admittedly I know just enough to be dangerous so perhaps someone else can give you a more definitive answer. -pickles Share this post Link to post Share on other sites More sharing options...
synergymus Posted July 25, 2005 Posted July 25, 2005 No it isn't overclocked. I do have the -forceasm switch on. I wonder if that is the problem. It folding all last week without a problem though. I t might be getting hot. I haven't check the temps on this box since I set it up as a server(since it wasn't doing anything terribly hard) but it is a athlonXP(barton)...and we all know about the temps on those suckers. Share this post Link to post Share on other sites More sharing options...
Guest area51 Posted July 25, 2005 Posted July 25, 2005 Look like we will break 2000 today or tomorrow... Does anyone have a good understanding of the wu's and score's of this project? I looked on their page and saw that they use a benchmark comparison w/a 2.8 P4 to generate scores...but the scores seem so inconsistent. Does anyone have any insight here? What is the effect of taking (or not) deadline wu's for example? Why do I get way more points for small wu's than big ones? Share this post Link to post Share on other sites More sharing options...
Angry_Games Posted July 25, 2005 Posted July 25, 2005 area51 i dont even worry about that stuff...my rigs are like the Popiel Pocket Fisherman...set it and forget it good ol 2640Mhz X2 4400+ is just kicking butt right now heh i finally got a damn 3200+ venice cpu also... Share this post Link to post Share on other sites More sharing options...
Guest area51 Posted July 25, 2005 Posted July 25, 2005 Suspect you are right AG...just curious. I am about to get my new 24/7 linux server into the folding game here in a couple days or so after I deal with more pressing configuration issues. It should help us out quite nicely. On a side note...went and looked at the #1 team (Overclockers Australia I think)...they got half their country's population on that damn team! Check out how many members they have! Got to give them their props...admirable effort! Share this post Link to post Share on other sites More sharing options...
seamuso Posted July 25, 2005 Posted July 25, 2005 oh my .. turned this flag on .. -advmethods ... just got a 10000 frame gromacs .. heh Share this post Link to post Share on other sites More sharing options...
synergymus Posted July 26, 2005 Posted July 26, 2005 area51, check out the official folding forums for tons of discussion on the points system. From what I have read the inconsisentencies come from serveral issues. First off, the performace of Intel, AMD, PowerPC, with the code varies greatly. AMD's do really well on some cores that Intel can't handle and vice-versa. Part of this has to do with SSE2 libraries which are Intels and supposedly AMD can't take advantage of them. The later doesn't matter though since the code doesn't recognize AMD64, Pentum M, etc, etc. (AMD64 is recognized as a AthlonXP, Pentum M as PIII) This goes on and on and on!!!!! Share this post Link to post Share on other sites More sharing options...
seamuso Posted July 26, 2005 Posted July 26, 2005 [19:43:11] Protein: p1549_A21nat_d13_GO [19:43:11] [19:43:11] Writing local files [19:43:11] Extra SSE boost OK. If you use the -advmethods flag on the startup shortcut or in command line for linux, you end up getting the gromacs + sse boost Share this post Link to post Share on other sites More sharing options...
cpuz Posted July 26, 2005 Posted July 26, 2005 OMG, we are really rockin n rollin, just broke 2,000....keep it up people! good work!!! Share this post Link to post Share on other sites More sharing options...
synergymus Posted July 26, 2005 Posted July 26, 2005 OMG, we are really rockin n rollin, just broke 2,000....keep it up people! good work!!! Great job everyone. Unfortunately my main rigs weren't contributing last night. I went to bed early and my wife shut everything down. Even though I told her those computers were busy curing disease! Share this post Link to post Share on other sites More sharing options...
synergymus Posted July 26, 2005 Posted July 26, 2005 [19:43:11] Protein: p1549_A21nat_d13_GO[19:43:11] [19:43:11] Writing local files [19:43:11] Extra SSE boost OK. If you use the -advmethods flag on the startup shortcut or in command line for linux, you end up getting the gromacs + sse boost THats exactly why I have it on. I saw that the client wasn't detecting my SSE capabilities in Linux unless I had it on. With this on my linux box (athlonXp) gets 600 pt gromacs more consistently than my Athlon64 Share this post Link to post Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now